The research of visual attention mechanism model fuse multi-feature
Zhenkun Wen, Yi-hua Du, Huisi Wu, Lei Wang · 2014
With the rapid development of network technology and multimedia technology, the researchers found that the neurobiology of human vision, computer vision and image and video processing and efficient combination for image, video can provide a more good solution of content retrieval applications, on the one hand, the simulation and study of visual attention mechanisms, can accurate images, video scene division of the region; we analyze the current theoretical basis of mechanism model, Proposed a significant degree measure algorithm fused color significant information, density significant and frequency domain transform significant base on visual attention mechanism model focus on multi-feature integration. Make records of significant area combined with a significant figure and as foundation for subsequent research and work.